[PATCH 09/17] math: Use exp10m1f from CORE-MATH

[Paul Zimmermann]

       Hi,

> >>  Function: "exp10m1":
> >>  double: 4
> >> -float: 2
> >>  ldouble: 3
> > 
> > Does this mean "we're not testing it any more" or "we expect zero ULPs
> > and verify that" ?  I don't see any ": 0" in any of the ulps files so I
> > assume the latter.
> 
> The former, we are expecting a correctly rounded implementation for the
> inputs we test.

I guess Adhemerval meant the latter: the CORE-MATH code should be correctly
rounded for all inputs (and all rounding modes).

Paul